| 06:02 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 08:29 |
|
mmorgan joined #evergreen |
| 08:57 |
|
_adb joined #evergreen |
| 09:05 |
|
Dyrcona joined #evergreen |
| 11:18 |
pinesol_green |
csharp: [evergreen|Kyle Huckins] LP#1511358 Patron Survey Interface - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=e2ee72f> |
| 11:18 |
csharp |
bug 1511358 |
| 11:18 |
pinesol_green |
Launchpad bug 1511358 in Evergreen "webclient: Surveys are not accessible from patron account" [Medium,Fix released] https://launchpad.net/bugs/1511358 |
| 11:20 |
Dyrcona |
csharp: You know how to write a small Perl program to make those calls? If so, that would be a good way to test changes to that code. |
| 11:21 |
csharp |
Dyrcona: I don't, unfortunately :-/ |
| 11:25 |
Dyrcona |
I can paste something for you in a bit. Got something else to do right now. |
| 11:25 |
csharp |
Dyrcona: that would be great! - thank you |
| 11:44 |
jeff |
helps if i remember to start the survey. |
| 11:45 |
csharp |
jeff: and I think a quirk of surveys is that you have to start them the next day (i.e., not today) |
| 12:09 |
pastebot |
"Dyrcona" at 64.57.241.14 pasted "For csharp: pcrud script." (55 lines) at http://paste.evergreen-ils.org/908 |
| 12:12 |
Dyrcona |
And, that was an obvious edit of a script I wrote earlier this week to test something in open-ils.reporter. |
| 12:12 |
csharp |
Dyrcona++ # will test :-) |
| 12:13 |
Dyrcona |
I would have changed the $reporter variable to $request if I'd noticed. :) |
| 12:14 |
Dyrcona |
I haven't run that script, either, so beware typos. |
| 12:14 |
jeff |
confirmed in webby, it's fleshing far too much. |
| 13:07 |
Bmagic |
I agree with you Dyrcona |
| 13:36 |
|
khuckins__ joined #evergreen |
| 13:59 |
jeff |
there are days when i think i'm getting better at using jq, and there are days when i just give in and supplement what i know with a blunt usage of grep. |
| 14:02 |
csharp |
jeff: empty in the user editor - unfortunately there's no "we've already answered this so don't ask again" feature |
| 14:19 |
csharp |
btw, the lack of such a feature means that required surveys require an answer every time you edit and save the record |
| 14:19 |
csharp |
but that's a bug report for another day |
| 15:02 |
csharp |
I was just thinking about the fact that concerto doesn't approximate "real" |
| 15:02 |
csharp |
or realistic data scale |
| 15:03 |
csharp |
I think it would be beneficial to host a more realistic dataset somewhere that people can download for testing larger things |
| 15:03 |
jeff |
*nod* |
| 15:04 |
* csharp |
isn't expressing himself well |
| 15:04 |
jeff |
and/or generators. |
| 15:04 |
csharp |
anyway - keep concerto around for basic proof of concept, but when doing testing for signoffs, require the larger dataset |
| 15:04 |
csharp |
just thinking out loud |
| 15:05 |
* mmorgan |
agrees. Many times production level data makes all the difference. |
| 15:06 |
Dyrcona |
Yeah. It does. |
| 15:08 |
kmlussier |
csharp: There are several projects that I will only test using something closer to production data. For example, the recent 3.0 search work was tested against a much larger dataset. |
| 15:08 |
kmlussier |
Testing search against Concerto is never a good idea. |
| 15:08 |
csharp |
kmlussier: that's good to hear |
| 15:09 |
csharp |
well, I'm very happy we test extensively before go-live |
| 15:09 |
csharp |
this probably would have brought us to our knees on day one |
| 15:10 |
csharp |
fwiw, the offending code is also in 2.12 - which tells me that not many are using the web client in production yet |
| 15:10 |
kmlussier |
csharp: Maybe, but do you have a good sense of how many people are using surveys? |
| 15:11 |
csharp |
kmlussier: everyone in PINES uses surveys - we require it to record voter registration |
| 15:12 |
Dyrcona |
You can register to vote at the library? |
| 15:20 |
|
Jillianne joined #evergreen |
| 17:06 |
|
mmorgan left #evergreen |
| 17:45 |
|
jvwoolf left #evergreen |
| 18:02 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 21:32 |
|
finnx1 joined #evergreen |
| 21:36 |
|
finnx1 left #evergreen |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:04 |
|
agoben joined #evergreen |
| 07:07 |
|
JBoyer joined #evergreen |
| 07:53 |
|
dwgreen joined #evergreen |
| 08:39 |
|
mmorgan joined #evergreen |
| 09:08 |
|
_adb joined #evergreen |
| 09:17 |
|
Dyrcona joined #evergreen |
| 09:24 |
csharp |
hmm - something on our 3.0.0 test server is blasting the DB with perm checks for permission.usr_has_object_perm(9554, 'VIEW_USER', 'asvr', '1541810') AS has_perm; |
| 09:24 |
csharp |
and it appears to be walking through every survey response |
| 09:24 |
csharp |
it has exhausted pcrud drones on all servers |
| 09:25 |
csharp |
many checks per second |
| 09:25 |
* csharp |
thinks something somewhere is coded all wrong |
| 09:27 |
csharp |
having trouble recreating it |
| 09:28 |
csharp |
yeah - this is a problem |
| 09:28 |
csharp |
it's happened to multiple users over the last few days |
| 09:31 |
csharp |
aha - it happens when I go to Other -> Surveys in a patron record |
| 09:34 |
csharp |
and yep, it's walking through every survey response in the DB |
| 09:35 |
csharp |
we have 3,895,615 survey responses to walk through |
| 09:35 |
* csharp |
wonders if JBoyer or Dyrcona has seen this |
| 09:35 |
csharp |
but I'm not sure if anyone else uses surveys to the same extent as we do |
| 09:36 |
JBoyer |
We have never used the Survey feature here, so no. (and thank goodness...) |
| 09:36 |
csharp |
JBoyer: yeah, thank your lucky stars! |
| 09:37 |
Dyrcona |
I don't think we use the survey feature either. We're still on 2.12, anyway. |
| 09:37 |
JBoyer |
But yes, it should obviously only do the check once, and I would assume by the point you've reached that interface it's been done already anyway. |
| 09:37 |
csharp |
Dyrcona: oh - I forgot about that |
| 09:37 |
Dyrcona |
And, I've not finished the 3.0 test server, yet. |
| 09:37 |
JBoyer |
\me is the only sucker that upgraded week 1. |
| 09:38 |
* JBoyer |
too |
| 09:38 |
csharp |
JBoyer++ pioneering |
| 14:25 |
|
khuckins__ joined #evergreen |
| 14:27 |
jeff |
JBoyer: based on last year, could be handy. |
| 14:28 |
JBoyer |
jeff++ # can't hurt to be prepared in any case. |
| 14:42 |
pinesol_green |
[evergreen|Remington Steed] Docs: Fix AsciiDoc header syntax bug - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=24c7160> |
| 15:02 |
|
sandbergja joined #evergreen |
| 15:18 |
|
mmorgan1 joined #evergreen |
| 15:21 |
|
khuckins_ joined #evergreen |
| 17:12 |
|
jvwoolf left #evergreen |
| 17:21 |
|
mmorgan left #evergreen |
| 17:58 |
|
Jillianne joined #evergreen |
| 18:02 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 01:01 |
|
tsbere_ joined #evergreen |
| 03:06 |
|
tsbere__ joined #evergreen |
| 04:52 |
|
jeff__ joined #evergreen |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 06:56 |
|
dwgreen joined #evergreen |
| 07:19 |
|
kmlussier joined #evergreen |
| 08:04 |
|
collum joined #evergreen |
| 10:05 |
berick |
not saying Perl's not to blame, just that all we know for sure at this point is jabber is mad, maybe at some xml the perl code is generating that's not valid. |
| 10:06 |
berick |
another option is to set loglevel to 5 -- that will show the XML we're sending to the jabber server |
| 10:06 |
berick |
(and it will slow everything down, beware) |
| 10:06 |
Dyrcona |
It's a test vm. No one else uses it. |
| 10:06 |
berick |
you could also up the log level, but only restart trigger |
| 10:07 |
berick |
so you're not slowing /everything/ down |
| 10:07 |
csharp |
berick: ooh - that's a good idea - I never thought of handling debug that way |
| 13:16 |
Dyrcona |
<username>opensrf</username><password>password</password> |
| 13:17 |
Dyrcona |
There's my 'super secret' password in the clear. |
| 13:17 |
bshum |
Fun |
| 13:17 |
Dyrcona |
For the logs, I'm lazy and just use password on my test vms because they're behind a firewall. |
| 13:18 |
Dyrcona |
There is a lot of "unnecessary" chatter going on. |
| 13:19 |
csharp |
Dyrcona: ah - good to know |
| 13:19 |
Dyrcona |
<<"orcester Tatnuck Magnet School\\\", |
| 13:34 |
Dyrcona |
So, I think I'll upload this 6MB of osrfsys.log to the Lp bug. I think miker had asked me to share it if I could. |
| 13:35 |
Dyrcona |
It is surprisingly small consider the log settings. |
| 13:35 |
Dyrcona |
Or, should i set loglength to the same as my ejabberd max_stana_size so the message that is too big will be clearly truncated? |
| 13:41 |
kmlussier |
I am testing something with floating, but I'm unable to get the basic floating behavior to work. I think I may be missing a step. |
| 13:41 |
kmlussier |
I edited a copy to use the default 'Everywhere' rule. If I check that copy in at another library, it should just stay there, right? Or do I need to do something else? |
| 13:42 |
kmlussier |
It keeps putting the copy into transit when I check it in at another location. :-/ |
| 13:43 |
|
jwoodard joined #evergreen |
| 13:47 |
kmlussier |
Hmmm...I can get it to work on my 2.12 server. Maybe there's a bug. |
| 13:48 |
* Dyrcona |
has never really looked into the floating implementation. |
| 13:54 |
kmlussier |
Oh, never mind. There was a hold on the title that the system was trying to fill. |
| 13:55 |
kmlussier |
This is what happens when I try substituting tea for coffee. |
| 17:01 |
|
jvwoolf left #evergreen |
| 17:09 |
|
mmorgan left #evergreen |
| 17:39 |
|
khuckins_ joined #evergreen |
| 18:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 08:12 |
|
Dyrcona joined #evergreen |
| 08:35 |
|
mmorgan joined #evergreen |
| 08:44 |
|
_adb joined #evergreen |
| 11:30 |
Dyrcona |
OK. |
| 11:42 |
miker |
Dyrcona: huh, yeah... the gateways are not chunking large messages |
| 11:43 |
* miker |
just looked at the new bug |
| 11:44 |
Dyrcona |
I have an acq issue that I think is the same thing. I can add the logs if you like, or wait to test a fix. |
| 11:45 |
* Dyrcona |
is still gathering the log information and likely won't have it all together until Monday, anyway. |
| 11:45 |
miker |
Dyrcona: if you can past the logs, that'd be good. I've got a small itch in the back of my mind re the router, now... |
| 11:46 |
miker |
and I certainly won't have a fix ready before monday... |
| 14:44 |
kmlussier |
hbrennan++ indeed |
| 14:45 |
csharp |
we broke down the options a few years ago: https://pines.georgialibraries.org/tip-pull-lists |
| 14:45 |
kmlussier |
hbrennan: Is it going fairly smoothly so far? |
| 14:46 |
hbrennan |
Thanks! |
| 14:46 |
hbrennan |
So far so good |
| 14:46 |
hbrennan |
I think things broke because we moved the whole database so we could test it out before. |
| 14:47 |
hbrennan |
Emails were being generated yesterday for courtesy notices, but getting stuck because some piece of Outlook was different |
| 14:47 |
hbrennan |
Just those types of issues |
| 14:47 |
|
mmorgan joined #evergreen |
| 14:47 |
hbrennan |
Staff (besides me) haven't had any trouble |
| 14:53 |
kmlussier |
hbrennan: Are people using the web client or are they still using the xul client? |
| 14:54 |
hbrennan |
Xul client for now. About half our staff is out this week |
| 14:54 |
hbrennan |
I haven't trained anyone on Webby yet |
| 16:53 |
kmlussier |
Have a nice weekend #evergreen! |
| 17:01 |
|
mmorgan left #evergreen |
| 17:25 |
|
jvwoolf left #evergreen |
| 18:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 19:30 |
|
khuckins_ joined #evergreen |
| 00:25 |
|
yar joined #evergreen |
| 03:50 |
|
gsams joined #evergreen |
| 06:00 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:12 |
|
rjackson_isl joined #evergreen |
| 07:40 |
|
rlefaive joined #evergreen |
| 07:54 |
|
kmlussier joined #evergreen |
| 12:57 |
Dyrcona |
Not seeing how I can force the join order with JSON queries, I'm going to try gmcharlt's squashed version of jeffdavis's branch. |
| 13:09 |
dbwells |
Dyrcona: Just curious, did you try bumping up the join_collapse_limit? |
| 13:09 |
Dyrcona |
No. I didn't. |
| 13:11 |
Dyrcona |
Trouble is. I don't have a comparable db to production to test with. My testing systems always seem slow. |
| 13:11 |
dbwells |
Ultimately that's all we did to work around it when bug 1527731 bit us. Not sure what our overall penalty spread for doing that has been, but things are fast enough. |
| 13:11 |
pinesol_green |
Launchpad bug 1527731 in OpenSRF "Query for OPAC copies can be extremely slow" [Undecided,New] https://launchpad.net/bugs/1527731 |
| 13:12 |
Dyrcona |
dbwells: What, exactly, did you set? |
| 13:19 |
dbwells |
Plus, I haven't followed closely whether your issue is intermittent or constant, so maybe it doesn't really apply. |
| 13:20 |
Dyrcona |
dbwells: It seems constant. |
| 13:20 |
Dyrcona |
Or, at least, constant enough. :) |
| 13:22 |
Dyrcona |
I'll try join_collapse_limit on my test db and see if it affects the plan on the slow version of the query. |
| 13:24 |
* dbwells |
tries to read up a bit more |
| 13:25 |
Dyrcona |
I can just "set join_collapse_limit = 9;" right? |
| 13:26 |
dbwells |
I think so. |
| 13:31 |
Dyrcona |
Just stating what I see for the logs, etc. |
| 13:34 |
Dyrcona |
dbwells++ |
| 13:35 |
dbwells |
Another interesting technique I learned recently is using join_collapse_limit = 1 to force the plan to execute the joins in the order given. Doesn't help when the root problem is not being able to control the requested join order at all, but could still come in handy. |
| 13:35 |
Dyrcona |
So, making that change on my test db server dropped the execution time of my "bad" query from 40-41 seconds to 13 milliseconds. |
| 13:35 |
Dyrcona |
dbwells: Yes, I saw that in the docs and in a stackoverflow question. |
| 13:36 |
Dyrcona |
And a mailing list post maybe.... |
| 13:36 |
dbwells |
Yeah, Erwin Brandstetter? Gotta love that guy :) |
| 17:25 |
kmlussier |
Bmagic: Unfortunately, I know very little about collections. |
| 17:26 |
Bmagic |
kmlussier: no problem, I think I just wanted to bounce it around here. I decided to say that "there isn't a way to merge them" - first you have to remove the patron from collections |
| 17:49 |
bshum |
dbwells++ # behind the scenes i18n detective work |
| 18:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 18:44 |
|
jvwoolf left #evergreen |
| 19:21 |
csharp |
Dyrcona: check auditor.actor_usr_history for the audit_usr for the weird patron edit |
| 19:22 |
Dyrcona |
csharp: I think that was meant for Bmagic. |
| 01:31 |
|
Jillianne joined #evergreen |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:43 |
|
dwgreen joined #evergreen |
| 08:05 |
|
kmlussier joined #evergreen |
| 08:09 |
|
collum joined #evergreen |
| 09:21 |
|
yboston joined #evergreen |
| 09:45 |
|
collum joined #evergreen |
| 09:50 |
csharp |
does webby respect the session timeout YAOUSes? |
| 09:52 |
mmorgan |
csharp: I don't believe I tested in webby explicitly, but see lp 1693035 |
| 09:52 |
pinesol_green |
Launchpad bug 1693035 in Evergreen "Logins not honoring all org unit timeout settings" [Medium,New] https://launchpad.net/bugs/1693035 |
| 09:55 |
csharp |
mmorgan: perfect - thanks! |
| 10:01 |
kmlussier |
csharp: I asked that question in here a few weeks ago. Initially, I was seeing that webby wasn't timing out at all, but the next day, it started to time out based on the YAOUS. |
| 10:02 |
kmlussier |
I didn't investigate further because other things came up. |
| 10:03 |
Dyrcona |
Webby is open. Maybe someone messed with the timeouts? |
| 10:04 |
kmlussier |
Yes, let me correct myself. Not in webby, but in the web client. I did the testing on my own VM, and it wasn't an issue with the settings changing. |
| 10:08 |
csharp |
sorry, yes I was using "webby" to mean "the web client" |
| 10:11 |
kmlussier |
csharp: Dyrcona was speculating that it could be a cache issue when I reported back my results here - http://irc.evergreen-ils.org/evergreen/2017-09-29#i_327518 |
| 10:12 |
kmlussier |
But a couple of days earlier when I first started testing, I believe berick said he noticed it doesn't always work for him. |
| 10:13 |
Dyrcona |
If there are errors rendering a template, where would they show up in the logs, if at all? |
| 10:14 |
berick |
i confirmed one scenario where it fails to log out. if the server is no longer responding, as in the case w/ a temp VM I set up yesterday, it get stuck in the middle of the log out dance. that's an atypical example, but maybe there's something to be learned there. worth looking for JS console errors around the time it should have logged out... |
| 10:15 |
Dyrcona |
Oh, wait! I see the problem. Never mind. |
| 10:40 |
|
mmorgan1 joined #evergreen |
| 10:41 |
csharp |
berick: not that I've seen |
| 10:44 |
csharp |
seeing lots of no_tz.open-ils.storage.actor.user.crazy_search: prepare_cached(SELECT evergreen.unaccent_and_squash(?)) statement handle DBIx::ContextualFetch::st=HASH(0xd55f878) still Active at /usr/local/share/perl/5.18.2/OpenILS/Application/Storage/Publisher/actor.pm line 627. in the osrfwarn log |
| 10:49 |
kmlussier |
berick: I didn't file any mainly because, once I got the timeout the work a couple of days after adding the setting, I didn't continue testing it to determine if my initial problem was indeed a cache issue or if there was a larger problem. |
| 10:51 |
berick |
kmlussier: gotcha. (again, not sure if we need one, just curious) |
| 10:51 |
* berick |
looks at the API issue |
| 10:53 |
jeff |
csharp: and did those insanely high values then cause problems with webby? |
| 11:16 |
|
sandbergja joined #evergreen |
| 11:31 |
Bmagic |
berick: select * from reporter.simple_record where tcn_value = '2468087' results in 4 rows. Perhaps too many rows is the issue? |
| 11:32 |
berick |
Bmagic: do where id = <whatever> |
| 11:32 |
Bmagic |
berick: that can't be right, because I tested one of the barcodes that showed the title in the email and it had 4 rows also |
| 11:33 |
Bmagic |
berick: using the id column to match the record from asset.call_number where call_number is asset.copy.call_number, I still get 4 rows |
| 11:34 |
berick |
ok, yeah, there should only ever be 1 reporter.simple_record entry per bib ID |
| 11:34 |
Bmagic |
strange that the example where the title appeared in the email also has 4 rows |
| 11:41 |
Bmagic |
funny you should ask, I did a presentation 2 years ago showing how we have been addressing duplicate records |
| 11:42 |
Bmagic |
and we have been deduping bre on a regular basis using this 2-step approach |
| 11:43 |
berick |
well i mean multiple occurences of the same bib ID in reporter.materialized_simple_record |
| 11:44 |
Bmagic |
berick: once I started querying rmsr - there is only 1 row per bib (of the two that I tested where one is showing the title and the other is not) |
| 11:45 |
berick |
ok, good. |
| 11:45 |
berick |
that's at leat as it should be |
| 11:45 |
Bmagic |
if there is a row there, then the email should have the title? |
| 15:03 |
miker |
it shouldn't, as the network should be autoflushed, and the send() def happens before the check for max_requests... but it's a lead I'm chasing down |
| 15:03 |
Dyrcona |
dbwells: Variation on the off-by-one error: Taking a count too soon and acting on it. |
| 15:07 |
dbwells |
miker: I am able to recreate the "hang" with enough persistence using open-ils.search.biblio.copy_counts.location.summary.retrieve but it doesn't look like the log message is related, unfortunately. |
| 15:07 |
miker |
:( |
| 15:07 |
miker |
thanks for testing |
| 15:10 |
miker |
dbwells: I was going to say, before, the stderr msg looks like a bug of omission, by not closing a helper statement handle that gets cached somewheres ... seems more likely now, after your test |
| 15:11 |
miker |
so, likely not a failure-causing bug, just noise |
| 15:11 |
dbwells |
yeah |
| 15:16 |
csharp |
this call: CALL: open-ils.search open-ils.search.serial.record.bib.retrieve 5621115, 1, 1 |
| 15:17 |
csharp |
which is made while a record loads, results in a "severe query error" (Empty IN list) |
| 16:13 |
mmorgan |
open-ils.cstore open-ils.cstore.direct.config.usr_setting_type.search.atomic {"name":[]} |
| 16:16 |
dbwells |
miker: Just caught up on reading bug #1704396. Not sure if my issue is even related to the other search hangs, but is sure seems that way. The "good" news is a simple srfsh script with 500 open-ils.search.biblio.copy_counts.location.summary.retrieve calls is enough to reliably trigger this behavior for us, unfortunately only on production. |
| 16:16 |
pinesol_green |
Launchpad bug 1704396 in Evergreen "Slowness for metarecord and one-hit searches in 2.12" [High,Confirmed] https://launchpad.net/bugs/1704396 |
| 16:16 |
dbwells |
miker: so, I should at least be in a somewhat reasonable position to test patches. |
| 16:19 |
dbwells |
miker: I also wonder (without much understanding yet) whether the bug might be in the ".atomic" wrapper. Logs look like storage responds, but search just keeps waiting for it to finish. I might try a quick non-atomic copy of open-ils.search.biblio.copy_counts.location.summary.retrieve just to see if it passes the srfsh test. |
| 16:30 |
|
b_bonner joined #evergreen |
| 16:35 |
|
rlefaive joined #evergreen |
| 16:43 |
dbwells |
Just anecdotal at this point, but "atomic" in the storage call does not seem to be a factor. |
| 17:05 |
miker |
dbwells: thanks! that's one less thing to check early on |
| 17:09 |
|
mmorgan left #evergreen |
| 17:11 |
|
Jillianne joined #evergreen |
| 17:12 |
|
jihpringle joined #evergreen |
| 17:50 |
|
b_bonner left #evergreen |
| 17:54 |
|
Dyrcona joined #evergreen |
| 18:02 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 18:23 |
|
phooks joined #evergreen |
| 19:26 |
|
phooka joined #evergreen |
| 19:30 |
phooka |
working on building out new production servers and autogen.sh is erroring out while Updating OrgTree |
| 00:13 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: adding backup information to cli manual - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=7ecad6e> |
| 02:27 |
|
kipd joined #evergreen |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:18 |
|
rjackson_isl joined #evergreen |
| 08:15 |
|
collum joined #evergreen |
| 08:43 |
|
mmorgan joined #evergreen |
| 09:01 |
JBoyer |
Or, anyone else with some insider knowledge re: 3.0 search. |
| 09:01 |
JBoyer |
? |
| 09:01 |
|
Dyrcona joined #evergreen |
| 09:05 |
* csharp |
is curious about JBoyer's search concern since we're testing 3.0 right now |
| 09:06 |
JBoyer |
I've found a difference in the queries where staff can't accurately limit to a single location but patrons can. I was hoping to get some background on why. |
| 09:06 |
JBoyer |
Also, has your 3.0 testing turned up anything similar? (or, rather, have you checked?) |
| 09:12 |
csharp |
not that I know of - I can check |
| 09:14 |
mmorgan |
JBoyer: I just did a quick search logged in as a staff user on a 3.0 system, and am also seeing a problem with limiting. |
| 09:15 |
JBoyer |
Ok. So good news, I'm neither crazy* nor running a damaged system. (*At least not for this reason) |
| 11:13 |
miker |
hrm... well, I'm around now |
| 11:15 |
miker |
re the search stuff above, I looked at this last week with an upgraded system and didn't see the behavior. which obv doesn't mean it's not happening, just that something's different |
| 11:19 |
|
mmorgan joined #evergreen |
| 11:19 |
kmlussier |
I see it too on one of my test VMs. |
| 11:30 |
|
rlefaive joined #evergreen |
| 11:46 |
kmlussier |
Never mind. I don't see it on my VM. I was just seeing the standard gray bibs that always come up in a staff search. |
| 11:46 |
* kmlussier |
drinks more coffee. |
| 11:47 |
kmlussier |
This dataset shouldn't have any copy-less bibs, but I guess that's another problem to solve in my spare time on a later date. |
| 12:02 |
|
khuckins joined #evergreen |
| 12:58 |
|
yboston joined #evergreen |
| 12:59 |
|
khuckins joined #evergreen |
| 17:48 |
csharp |
maybe the wrong kind of join in the fieldmapper linkage? (but that probably would result in perl-level errors rather than blank data) |
| 17:54 |
berick |
Bmagic: you've confirmed the bibs in question have reporter.simple_record entries? |
| 17:54 |
berick |
... with titles |
| 18:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 19:04 |
rhamby |
fyi, removing roles of some wordpress accounts for those that haven't been active in 8+ years |
| 19:12 |
csharp |
wow - that bot really was wreaking havoc for us - error/warn logs are very reasonable now |
| 19:12 |
csharp |
@band add Wreaking Havoc |
| 00:41 |
|
Jillianne joined #evergreen |
| 06:00 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:16 |
|
rjackson_isl_ joined #evergreen |
| 08:11 |
|
collum joined #evergreen |
| 08:27 |
|
kmlussier joined #evergreen |
| 09:19 |
|
bos20k joined #evergreen |
| 09:48 |
miker |
jeff / bshum: yes, there's really nothing to maintain -- the introspection stuff is old as the hills and hasn't changed ... ever. there are some issues on the C side that should be cleaned up (the atomic/streaming bool is logically reversed from the perl, signature structure is a little different) however -- but that's opensrf, not docgen |
| 09:49 |
miker |
also, I find it very handy on a dev system on occasion, so I'd like to see it unbroken as well |
| 09:55 |
bshum |
Well, so far in my re-testing of the apache config |
| 09:55 |
bshum |
I think all that's really needed was to add "AddType application/xml .xsl" to the definition |
| 09:55 |
bshum |
I'm not sure if it's an apache 2.4 specific thing |
| 09:56 |
bshum |
I have to retest with a system that's got apache 2.2, but really those should be mostly dead by now :) |
| 09:56 |
miker |
bshum: that looks like it works. 2.2 isn't all the way dead, though |
| 09:56 |
* bshum |
goes back to the archives to try finding a wheezy ISO |
| 09:57 |
bshum |
miker: I know, I'm just living my life on the edge of new :) |
| 10:53 |
bshum |
So apache 2.2 config requires no adjustment and "just works" |
| 10:54 |
miker |
it correctly sees xsl as an xml application ... :) |
| 10:54 |
jeff |
bshum: on apache 2.4, you didn't need the SSI legacy directive? That doesn't mesh with my understanding. |
| 10:54 |
bshum |
jeff: Well I was just testing with the lines one by one, and found that I only needed the one line |
| 10:55 |
miker |
jeff: you do. you need the whole block that the commit removed, plus the line bshum is talking about |
| 10:55 |
miker |
or, that's what worked for me on 2.4 |
| 10:55 |
bshum |
So now I'm not sure why, cause I thought I needed everything too |
| 10:55 |
bshum |
Maybe something got cached... re-tests |
| 10:55 |
jeff |
heh. |
| 10:56 |
jeff |
miker: that's why i was surprised and trying to make sure i understood bshum's statements above. |
| 10:57 |
bshum |
jeff: miker: well.... odd as it seems, when I only add that one line to my eg_vhost.conf definition for the docgen.xsl, the page renders as expected... |
| 16:17 |
|
mmorgan joined #evergreen |
| 17:10 |
|
mmorgan left #evergreen |
| 17:13 |
|
sandbergja joined #evergreen |
| 18:00 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 18:18 |
|
jvwoolf joined #evergreen |
| 06:01 |
|
rlefaive joined #evergreen |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:10 |
|
agoben joined #evergreen |
| 07:13 |
|
rjackson_isl joined #evergreen |
| 07:16 |
|
JBoyer joined #evergreen |
| 08:44 |
|
_adb joined #evergreen |
| 08:44 |
|
mmorgan joined #evergreen |
| 08:50 |
Dyrcona |
Based on this line "dumpost_cfg_t *) ap_get_module_config(r->per_dir_config, &dumpost_module)" I assume that mod_dumpost can be configured per directory, so I'll give that one a shot. It can also log headers. |
| 09:01 |
JBoyer |
Oh, yeah, dumpio was what I had used. I was using it on a test system so while it was crazy chatty it wasn't *All the things* I hope dumppost works, that sounds more generally useful. |
| 09:04 |
Dyrcona |
I'll give it a whirl. |
| 09:19 |
|
abowling1 joined #evergreen |
| 09:19 |
Dyrcona |
I have to use this on production to capture what Autographics sends us. I'd rather not have to take the time to set up a test with them. |
| 09:19 |
Dyrcona |
But, I'm gonna try it on a test vm, first. |
| 09:21 |
Dyrcona |
@decide stay or go |
| 09:21 |
pinesol_green |
Dyrcona: go with go |
| 09:22 |
Dyrcona |
pinesol_green: I knew you'd say that. ;) |
| 10:09 |
Dyrcona |
Ah, no. It looks like that line is deceiving and DumpPost configs have to be server wide. |
| 10:26 |
Dyrcona |
JBoyer: I've logged a few messages from Autographics, and I'm surprised. |
| 10:27 |
Dyrcona |
JBoyer: What I see blows up with the old version of HTTP::Body, too, so the change that causes the issue is elsewhere. |
| 10:27 |
Dyrcona |
But, now I have some real data to test with. |
| 10:34 |
JBoyer |
Dyrcona, did you manage to pull any POST info with them or just the XML? If it's not too much trouble to anonymize I could poke at it on occasional too |
| 10:34 |
Dyrcona |
I'm grabbing the content-type head and the POST body. |
| 10:34 |
Dyrcona |
s/(head)/\1er/ |
| 10:35 |
Dyrcona |
Since this is server-wide. |
| 10:36 |
Dyrcona |
I've got at least 5 more messages from autographics, now. |
| 10:37 |
Dyrcona |
I may shut it down before an hour. The file grows a few hundred KB per minute. :) |
| 10:39 |
Dyrcona |
They're using Content-type: application/xml, and my testing shows that blows with Dancer 1.3202 with the old and the new version of HTTP::Body. |
| 10:39 |
JBoyer |
Not as bad as dumpio but still a but chatty, yeah |
| 10:40 |
Dyrcona |
In fact, the xml handler for HTTP::Body is the same, other than the version number. |
| 10:40 |
JBoyer |
Huh. So something about the older version of HTTP::Body on 14.04 is why it doesn't blow up here? |
| 13:36 |
JBoyer |
And 2, when customizing receipt templates, (checkout specifically) I |
| 13:36 |
Dyrcona |
OK. It shouldn't be hard to recreate. Just set content-type to application/xml for one. |
| 13:37 |
JBoyer |
ve heard from staff that they can't add patron name and barcode. Meaning that {{patron.first_given_name}} and so on works in the preview, but not on paper. Is that a regression or is the preview just ahead of the backend? |
| 13:37 |
Dyrcona |
JBoyer might also be interested to know that after 2-1/2 hours of logging, I have 127 actual messages from autographics to test with. |
| 13:37 |
dbwells |
Dyrcona: yes :) In my opinion, the bug is in HTTP::Body::XForms. It just lazily sets body to the buffer contents, which isn't going to be a handle. |
| 13:38 |
Dyrcona |
dbwells: What's odd is that hasn't changed since 2010, and that older version "worked." |
| 13:38 |
Dyrcona |
Unless autographics changed their content-type to coincide with our upgrade... |
| 13:54 |
Dyrcona |
That may be easier. I'll try that first. |
| 13:54 |
Dyrcona |
dbwells++ # again for the suggestions. |
| 13:59 |
dbwells |
Dyrcona: trying to familiarize myself a bit with NCIPServer, figure we'll probably move in that direction sooner or later :) |
| 14:00 |
kmlussier |
JBoyer: You might want to check in with Terran on the receipt issue. I know she did a lot of testing to see which fields worked and which didn't. |
| 14:02 |
kmlussier |
JBoyer: When I tested the new search code, I did a lot of side-by-side testing on two different servers to try to verify the code was retrieving the same set of search results, except in cases where it was retrieving more because of the higher cap. |
| 14:02 |
kmlussier |
I didn't see anything strange at that time, but I don't know if I limited by library from advanced search in my testing. |
| 14:03 |
* kmlussier |
returns to trying to figure out why badges don't always show when they should. |
| 14:19 |
miker |
JBoyer: not seeing that on demo.evergreencatalog.com ... and I can't remember you're catalog's URL |
| 14:20 |
JBoyer |
evergreen.lib.in.us |
| 14:21 |
JBoyer |
berick, re: bug 1656036, If I have time tomorrow to through something together I'd appreciate some thoughts, but if I can't get things together go with your plan. Like you said, it's better to get something going than continually bikeshedding this thing. |
| 14:22 |
JBoyer |
Yeah, that metarecord search bug was driving me nuts, but I didn't see any point in letting people see it. |
| 14:22 |
JBoyer |
I mean, as far as I know it still is, I haven't had time to track anything down. :/ |
| 14:23 |
JBoyer |
miker, a simple search that shows what I mean: search Indiana state library for title Bag of Bones (we mainly collect large print for users with poor sight) there are several results that have nothing to do with us. |
| 14:26 |
miker |
so, I may not be testing what you're testing, but I just searched for 'harry potter' globally (773 results), went to advance search and restricted to Andrews and got what I expected (77 results) and added added a cache killer, and got the same 77 results. all hav copies at andrews AFAICT |
| 14:26 |
kmlussier |
JBoyer: They're all ebooks? |
| 14:26 |
miker |
or are ebooks |
| 14:26 |
|
khuckins_ joined #evergreen |
| 14:37 |
miker |
:) |
| 14:40 |
|
tspindler joined #evergreen |
| 14:42 |
JBoyer |
I'm currently rebuilding all of the app servers to hide the timings, but since it's only staff I guess it's less dire, though staff are more likely to notice anyway. |
| 14:42 |
JBoyer |
So I'll keep looking, and that's a way to test webby to make sure it's just us. |
| 14:48 |
Dyrcona |
dbwells++ # One more time, because hacking HTTP::Body::XForms seems to have fixed it for me. Changing NCIP::Dancing didn't work. :( |
| 14:50 |
miker |
JBoyer: I'm not seeing this in a (quick) test at demo.evergreencatalog.com, fwiw |
| 14:51 |
miker |
logged in as staff, obv |
| 14:51 |
JBoyer |
miker, thanks for checking; moar data, moar help. I'll continue to poke. |
| 14:52 |
JBoyer |
Just because I'm curious, does the new asset.copy_vis_cache stuff get used in searches like this, or is that used elsewhere? (record details or something else) |
| 14:52 |
miker |
it's used in searches |
| 17:14 |
jeff |
i'd still recommend not installing it by default, and perhaps even recommend restricting it, and give it a look-over, but part of why it has lasted as long as it has is that it doesn't require much in the way of upkeep -- it uses opensrf system calls to get methods and their signatures from the running system. |
| 17:26 |
|
Jillianne joined #evergreen |
| 17:59 |
|
yboston joined #evergreen |
| 18:02 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 19:01 |
|
Christineb joined #evergreen |
| 19:22 |
* csharp |
cranks up Bmagic's docker image |
| 20:09 |
csharp |
btw, my suggestion to use Fedora server/cockpit makes downloading/installing/running the docker image super easy |
| 01:35 |
|
rlefaive joined #evergreen |
| 04:02 |
|
Jillianne joined #evergreen |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 06:32 |
|
book` joined #evergreen |
| 07:01 |
|
jvwoolf joined #evergreen |
| 07:04 |
|
sard joined #evergreen |
| 11:13 |
_adb |
O_o |
| 11:14 |
_adb |
idk 'bout that, but i got cats https://i.imgur.com/g27CrKb.gifv |
| 11:15 |
kmlussier |
:) |
| 11:31 |
Dyrcona |
Eh, bummer. That patch hasn't fixed the problem in production. It worked on my test VM though. |
| 11:32 |
|
jvwoolf joined #evergreen |
| 11:36 |
Dyrcona |
Maybe it did, and it didn't kick in until the second apache2 reload? |
| 11:40 |
|
collum joined #evergreen |
| 15:22 |
Bmagic |
kernal/kernel |
| 15:24 |
csharp |
sandbergja: also, make sure the laptop's processor is set to use virtualization (BIOS setting), otherwise vbox will not run well |
| 15:25 |
csharp |
(might be the case with Hyper-V too, but I'd be surprised if it runs at all without that setting set already) |
| 15:25 |
Bmagic |
sandbergja: If I am understanding that you are using docker for Windows, you can test a basic Ubuntu image without using the Evergreen container. "docker run ubuntu -it" |
| 15:27 |
Bmagic |
Something tells me that it's not going to work due to some of the system calls being incompatible. the videos: https://vimeo.com/230985351 and https://vimeo.com/231611654 |
| 15:27 |
csharp |
Bmagic++ |
| 15:27 |
* csharp |
still needs to make time to watch the third installment |
| 17:43 |
cesardv |
lol |
| 17:43 |
* cesardv |
really likes halloween >< |
| 17:44 |
* berick |
will not be trickrtreating at cesardv's house |
| 18:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 18:46 |
|
jvwoolf joined #evergreen |
| 18:59 |
|
stephengwills left #evergreen |
| 19:29 |
|
rlefaive joined #evergreen |
| 00:06 |
|
sallyf joined #evergreen |
| 00:06 |
|
lbarry joined #evergreen |
| 02:33 |
|
Jillianne joined #evergreen |
| 06:00 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:20 |
|
rjackson_isl joined #evergreen |
| 07:35 |
|
JBoyer joined #evergreen |
| 07:51 |
|
agoben joined #evergreen |
| 13:40 |
Dyrcona |
I noticed it when looking at the latest version of the Enlightenment window manager. They apparently use it. |
| 13:41 |
Dyrcona |
Right now, I'm trying to figure out why Plack::Handler::Apache2 and Dancer::Request are not playing nice together. |
| 13:55 |
Dyrcona |
So, I'm trying Plack from CPAN. |
| 14:02 |
Dyrcona |
No one can answer me about running NCIPServer on Ubuntu 16.04? It is broken for me, but I swear I tested is once before and it worked. |
| 14:21 |
Dyrcona |
Oh, lovely. |
| 14:21 |
Dyrcona |
It's broken in production, but it just works on a test vm. |
| 14:41 |
Dyrcona |
And, it just works for me in production when I run my test script using LWP against it. |
| 14:41 |
jeff |
problem with the data production is receiving? |
| 14:42 |
|
stephengwills joined #evergreen |
| 14:51 |
Dyrcona |
Nope. Not the data. |
| 15:29 |
jeff |
on git.evergreen-ils.org or elsewhere? |
| 15:30 |
Dyrcona |
git.evergreen-ils.org |
| 15:30 |
Dyrcona |
I've been chatting with some folks in #dancer on irc.perl.org. |
| 15:30 |
Dyrcona |
The consensus there was some weird interaction of Plack::Handler::Apache2 and Dancer, but that was before I had done my own tests. |
| 15:31 |
Dyrcona |
I even removed the packaged plack and installed it from CPAN. |
| 15:32 |
jeff |
did it break suddenly, or is this after a known change was made? |
| 15:32 |
Dyrcona |
Now, it think it's something "weird" in their POST requests that Apache 2.4 or the Plack handler doesn't like. |
| 17:12 |
jeff |
stephengwills: are you trying to step through a trigger function, or do something else? |
| 17:12 |
stephengwills |
stepping through trigger function |
| 17:13 |
stephengwills |
still working on that 001/003 thing from this morning. everytime I try to get out…it pulls me back in! |
| 17:13 |
Bmagic |
stephengwills: I have been known to replace a function with one that puts more in the output using things like NOTICE along the way. On a test machine of course. |
| 17:14 |
stephengwills |
I was thinking of doing that. just lots of watchdog lines |
| 17:14 |
Bmagic |
in some cases, adding columns to tables to write the debug output during execution. Maybe jeff has something better |
| 17:17 |
|
Jillianne2 joined #evergreen |
| 17:24 |
jeff |
if they don't, then they will fall back to org unit id 1, and use either that org unit's setting for cat.marc_control_number_identifier or they will use the shortname if org unit 1 does not have a value set for that setting. |
| 17:25 |
stephengwills |
so what i really want to do is run the bre and update the owner? on save the 003 will get fixed? |
| 17:25 |
jeff |
where "they will fall back to org unit id 1" is better written as "then the control number identifier handling for those records will fall back to org unit id 1" |
| 17:25 |
stephengwills |
that’s easy enough to test |
| 17:26 |
jeff |
the global flag cat.maintain_control_numbers also needs to be enabled, but i suspect you have that set already. |
| 17:27 |
stephengwills |
yup that’s good |
| 17:27 |
stephengwills |
almost all of our records have a null owner |
| 17:27 |
stephengwills |
:/ |
| 17:28 |
stephengwills |
you understand I only come here to embarress myself in from of my better, right? |
| 17:29 |
stephengwills |
thanks |
| 18:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 18:11 |
|
stephengwills left #evergreen |
| 21:18 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: updating copy buckets docs - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=ec21eb8> |
| 23:14 |
pinesol_green |
[evergreen|Jane Sandberg] Docs: documentation for Record Buckets - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=55edc1f> |
| 00:56 |
|
bwicksall_ joined #evergreen |
| 01:34 |
|
bwicksall joined #evergreen |
| 06:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:13 |
|
rjackson_isl joined #evergreen |
| 07:37 |
|
Dyrcona joined #evergreen |
| 07:53 |
|
kmlussier joined #evergreen |
| 08:11 |
Dyrcona |
Noodle [from Gorillaz] gets a holiday? Cool! |
| 08:11 |
jonadab |
Only thing is, if you have a Happy Puppy Day, you're probably going to have to have a Cute Kitten Day too, or the cat lovers will riot. |
| 08:12 |
rjackson_isl |
True |
| 08:12 |
kmlussier |
Dyrcona++ #Excellent test plan for placing multiple holds |
| 08:12 |
Dyrcona |
kmlussier: Thanks. There are also PgTap and perl tests for the backend changes. |
| 08:13 |
* Dyrcona |
has decided that tests are good. |
| 08:16 |
* csharp |
had an idea to develop baseline schema pgtap tests so sites know if they have everything expected by the stock install |
| 08:16 |
csharp |
but seems like that would be a massive project |
| 08:16 |
csharp |
actually a lot of the scripting could probably be automated |
| 08:16 |
Dyrcona |
Some of that is there already, I think. |
| 08:16 |
Dyrcona |
And, yes, it could be automated. |
| 08:17 |
Dyrcona |
Query pg_catalog and related tables on a known good installation and write out the tests. |
| 08:18 |
Dyrcona |
I still wouldn't call it trivial. |
| 08:18 |
Dyrcona |
Might take a day or so to get it all just right. |
| 08:18 |
* csharp |
keeps thinking of things like this, thinks, "oh, that would be great for the hackaway", then realizes how many other things exactly like that have already been mentioned :-) |
| 08:36 |
|
collum joined #evergreen |
| 08:52 |
|
kmlussier joined #evergreen |
| 09:00 |
|
bos20k joined #evergreen |
| 09:16 |
* Dyrcona |
shut down his gitlab test vm, but it may only be temporary. |
| 09:16 |
Dyrcona |
I might rebuild it or move it to another server. |
| 09:17 |
csharp |
gitlab is the front runner imho |
| 09:17 |
Dyrcona |
I really need to make the time to look at it seriously. I think I'd like to switch to gitlab from gitolite for our local repositories, but other projects are more important. |
| 09:26 |
|
kmlussier joined #evergreen |
| 09:27 |
csharp |
Dyrcona: I have another server earmarked for community VMs that has more resources than current mundungus |
| 09:27 |
Dyrcona |
csharp++ |
| 09:30 |
Dyrcona |
My gitlab test vm is set to use up to 60GB of space and is actually 18GB in size, and it's hardly being used. |
| 09:31 |
Dyrcona |
It uses 4GB of RAM, but could probably use more. |
| 09:31 |
Dyrcona |
Just for a reference. |
| 09:31 |
csharp |
/dev/mapper/mundungus-root 481G 257G 200G 57% / |
| 16:52 |
Dyrcona |
Not related (directly) to Evergreen but this looks neat: http://resume.github.io/ |
| 16:53 |
Dyrcona |
You have to opt-in by starring the resume/resume.github.com project. |
| 17:21 |
|
roycroft joined #evergreen |
| 18:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
